Qlist.ai contributed to engineering on a visual field testing platform for ophthalmic care — a healthcare system where technicians configure assessments, doctors monitor them live, and patients are tested on head-mounted display devices, with sub-second updates from clinic floor to oversight screen.
Visual field testing platform
Qlist.ai was engaged on core systems within the broader product — not as sole owner of design, build, and deployment of the complete platform. The summary below reflects the stack and workflows we helped implement and document.
Outcomes
- Real-time assessment updates from MongoDB Change Streams to doctor monitoring UI via WebSocket and SSE.
- Role-separated applications: technician configuration vs. doctor live oversight.
- Central NestJS API with Auth0 JWT validation and Redis-backed session caching.
- Assessment lifecycle from calibration through visual field testing, pause/resume, and completion.
- gzip mongodump backup pipeline to AWS S3 with local fallback and scheduled restore operations.
Challenge
The client is building a more accurate, reliable visual field testing system for glaucoma care. The workflow spans three clients — technician UI, doctor monitoring cube, and HMD devices for patients — requiring strict separation of duties, low-latency clinical visibility, and durable persistence with audit-friendly data handling.
Solution
- Contributed to the central API: versioned REST, Socket.io rooms, SSE channels, and change-stream watchers on assessment and HMD collections.
- Implemented auth integration — Auth0 OAuth/OIDC, JWKS validation, refresh tokens, and Redis session TTL aligned to token lifetime.
- Supported assessment and HMD domain flows: status transitions (calibration → VFT in progress → completed), device assignment, and clinical notes.
- Helped shape embedded MongoDB schemas for configuration and per-assessment result data, plus soft-delete patterns for audit trails.
- Documented backup/restore operations (mongodump → S3, mongorestore) and Docker/ECR deployment patterns for clinic installs.
Stack (platform)
- Central API — NestJS, MongoDB replica set, Redis, Socket.io, AWS S3, Python PDF/HVF service.
- Technician app — Next.js 14, Material-UI, React Query (patients, worklists, HMD admin, backups).
- Doctor monitoring app — Next.js 14, Material-UI, EventSource/SSE (live monitoring, eye-relief, summaries, notes).
Building regulated or real-time systems?
We help teams architect backends, ship clinical workflows, and document operations for production handoff.
Book a Call